NeurobiologyofDisease PI3K-AktSignalingActivatesmTOR ...traumatic epileptogenesis remain largely unknown (McDaniel and Wong, 2011), and effects of mTOR inhibitors on epilepsy have

Neurobiology of Disease

PI3K-Akt Signaling Activates mTOR-MediatedEpileptogenesis in Organotypic Hippocampal CultureModel of Post-Traumatic Epilepsy

Yevgeny Berdichevsky,1,2 Alexandra M. Dryer,3 Yero Saponjian,1 Mark M. Mahoney,4 Corrin A. Pimentel,4Corrina A. Lucini,4 Marija Usenovic,1 and Kevin J. Staley1,21Department of Neurology, Massachusetts General Hospital, Boston, Massachusetts 02129, 2Harvard Medical School, Boston, Massachusetts 02129, and3Department of Electrical and Computer Engineering and 4Bioengineering Program, Lehigh University, Bethlehem, Pennsylvania 18015

mTOR is activated in epilepsy, but the mechanisms of mTOR activation in post-traumatic epileptogenesis are unknown. It is also not clearwhether mTOR inhibition has an anti-epileptogenic, or merely anticonvulsive effect. The rat hippocampal organotypic culture model ofpost-traumatic epilepsy was used to study the effects of long-term (four weeks) inhibition of signaling pathways that interact with mTOR.Ictal activity was quantified by measurement of lactate production and electrical recordings, and cell death was quantified with lactatedehydrogenase (LDH) release measurements and Nissl-stained neuron counts. Lactate and LDH measurements were well correlated withelectrographic activity and neuron counts, respectively. Inhibition of PI3K and Akt prevented activation of mTOR, and was as effective asinhibition of mTOR in reducing ictal activity and cell death. A dual inhibitor of PI3K and mTOR, NVP-BEZ235, was also effective.Inhibition of mTOR with rapamycin reduced axon sprouting. Late start of rapamycin treatment was effective in reducing epileptic activityand cell death, while early termination of rapamycin treatment did not result in increased epileptic activity or cell death. The conclusionsof the study are as follows: (1) the organotypic hippocampal culture model of post-traumatic epilepsy comprises a rapid assay ofanti-epileptogenic and neuroprotective activities and, in this model (2) mTOR activation depends on PI3K-Akt signaling, and (3) tran-sient inhibition of mTOR has sustained effects on epilepsy.

IntroductionEpileptogenesis, or the development of epilepsy, after brain in-jury is characterized by a gradual and continual increase in sei-zure probability (Salazar et al., 1985; Williams et al., 2009;Raymont et al., 2010). One of the characteristics of the epilepticbrain is axon sprouting that has been best described at the mossyfiber pathway in the hippocampus (Cronin and Dudek, 1988;Sutula et al., 1989; Okazaki et al., 1995).

The mTOR pathway is transiently activated after experimentalepileptogenic brain injuries (Buckmaster et al., 2009; Zeng et al.,2009; Huang et al., 2010) and in the genetic syndrome of tuberoussclerosis, in which intractable epilepsy is a prominent feature(Crino, 2011; McDaniel and Wong, 2011; Staley et al., 2011).mTOR inhibition suppresses mossy fiber sprouting in models ofacquired epilepsy (Buckmaster et al., 2009) and reduces sponta-neous seizures in the kainate model of chronic epilepsy (Zeng et

al., 2009; Huang et al., 2010). Anti-epileptogenic effects weremore variable in the pilocarpine model (Buckmaster et al., 2009;Tang et al., 2012) and the acute kindling model (Sliwa et al., 2012;van Vliet et al., 2012).

Unlike traditional anticonvulsants, rapamycin has no imme-diate effects on seizures but rather alters hippocampal circuitry(McDaniel and Wong, 2011). mTOR inhibition could prevent orreverse post-traumatic changes in brain circuitry that lead to anepileptic state (Crino, 2011; Lew and Buckmaster, 2011; McDan-iel and Wong, 2011). However, rapamycin, the drug that is widelyused to inhibit mTOR, is a potent immunosuppressant (Chuehand Kahan, 2005), and the mTOR pathway has a broad role inregulation of protein synthesis and other growth-related cellularprocesses (Sarbassov et al., 2005; Caron et al., 2010). To developmore specific treatments, a better understanding of mTOR acti-vation in epilepsy is needed.

The mTOR pathway has been extensively studied in cancer(Chiang and Abraham, 2007; Guertin and Sabatini, 2007; Engel-man, 2009; Borders et al., 2010). Among regulators of mTORcontaining, rapamycin-sensitive protein complex (mTORC1) areRaf kinase (RAF) mitogen-activated protein kinase kinase (MEK)—ERK (RAF-MEK-ERK) and phosphatidylinositol 3-kinase (PI3K)—Akt pathways, and metabolite availability sensed through pathwaysinvolving AMPK, REDD1, and others.

PI3K signaling has recently been found to be necessary formTOR activation by pentylenetetrazole-induced seizures (Zhangand Wong, 2012), but upstream activators of mTOR in post-

Received Aug. 2, 2012; revised April 1, 2013; accepted April 8, 2013.Author contributions: Y.B. and K.J.S. designed research; Y.B., A.M.D., Y.S., M.M.M., C.A.P., C.A.L., and M.U.

performed research; Y.B. and K.J.S. analyzed data; Y.B. and K.J.S. wrote the paper.This study was supported by National Institutes of Health Grants R21 NS072258 and R01 NS077908 (National

Institute of Neurological Disorders and Stroke).Correspondence should be addressed to Kevin J. Staley, MGH Neurology, CNY-114, Room 2600, 114 16th Street,

Charlestown, MA 02129. E-mail: [email protected]. Berdichevsky’s present address: Bioengineering Program and Department of Electrical and Computer Engi-

neering, Lehigh University, Bethlehem, PA 18015.DOI:10.1523/JNEUROSCI.3870-12.2013

Copyright © 2013 the authors 0270-6474/13/339056-12$15.00/0

9056 • The Journal of Neuroscience, May 22, 2013 • 33(21):9056 –9067

traumatic epileptogenesis remain largely unknown (McDanieland Wong, 2011), and effects of mTOR inhibitors on epilepsyhave varied depending on the timing of intervention and themodel. To study mTOR signaling in epilepsy, we used an accel-erated in vitro model of post-traumatic epilepsy that capturescritical features of clinical epileptogenesis, including gradual on-set, axon sprouting, spontaneous electrographic seizures, seizureclustering, status epilepticus, and subsequent activity-dependentcell death, response to anticonvulsants, and emergence of anti-convulsant resistance (Dyhrfjeld-Johnsen et al., 2010; Berdi-chevsky et al., 2012). We developed a novel assay of epilepticactivity based on seizure-dependent increases in extracellular lac-tate concentration (During et al., 1994) to quantify epilepsy inthis preparation, and combined it with estimates of cell deathbased on release of the enzyme lactate dehydrogenase (LDH).

Materials and MethodsOrganotypic cultures. We generated organotypic hippocampal slice cul-tures from postnatal day 7 or 8 male Sprague Dawley rats, and main-tained them in poly-D-lysine-coated, 6-well tissue culture plates on arocking platform at �1 cycle/min, in 37°C and 5% CO2 in Neurobasal-A/B27 (Invitrogen, prepared according to manufacturer’s specifica-tions), with 0.5 mM GlutaMAX and 30 �g/ml gentamicin (Invitrogen).All animal use protocols were approved by the Institutional Animal Careand Use Committee at Massachusetts General Hospital and were con-ducted in accordance with the United States Public Health Service Policyon Humane Care and Use of Laboratory Animals.

Drugs and lactate and LDH assays. Rapamycin, SP 600125, U0126, andLY 294002 were purchased from Tocris Bioscience, and NVP-BEZ235was purchased from BioVision. All drugs were dissolved in dimethylsulf-oxide (DMSO) and applied to cultures starting on 3 d in vitro (DIV),unless otherwise specified. Control cultures were treated with 0.1%DMSO as vehicle. Since organotypic cultures from different animals de-velop epilepsy at slightly different rates, all experiments used controlslices from the same animal. Lactate and LDH concentrations were mea-sured in used culture medium with kits from BioVision. Lactate concen-trations were calculated relative to known lactate standards (BioVision),while LDH concentrations were calculated in terms of arbitrary units(a.u.), normalized to the 6 –28 DIV average of LDH concentration incontrol (epileptic) culture supernatant.

Electrophysiology and propidium iodide staining. Activity was recordedwith tungsten microelectrodes in CA1 pyramidal layer while cultureswere maintained in an interface chamber perfused with Neurobasal-A at37°C. Chronic multiple electrode array (MEA) recordings were per-formed with custom-designed chips (Berdichevsky et al., 2012), onwhich organotypic cultures were maintained as described above. Lactateconcentrations in the culture supernatant after 3 or 4 d in the incubatorwere compared with MEA electrical recordings from the same cultures.Dead cell stain was performed using propidium iodide (PI) (Invitrogen;1 mg/ml stock) added to live cultures for a final 1:250 dilution. Cultureswere incubated for 1 h, imaged on an inverted confocal microscope(Zeiss), and analyzed as described previously (Berdichevsky et al., 2012).Numbers of dead cells were counted during medium changes between 6and 25 DIV, and compared with relative LDH concentrations in usedculture medium from the same cultures during the same time period.

Nissl staining. Cultures were fixed in 4% paraformaldehyde, perme-abilized for 1 h in 0.3% Triton-X on a rocking platform, and stained withPI (Invitrogen; 1 mg/ml stock, diluted 1:250 in PBS) for 5 h. Cultureswere mounted, and images were acquired with a 40� objective on aconfocal microscope (Leica) with appropriate laser wavelength and fil-ters. Optical stacks spanning the entire thickness of the slice were re-corded, with z-direction step size of 3 �m.

Immunohistochemistry. Cultures were fixed and permeabilized. Block-ing buffer (10% goat serum in 0.05% Triton-X in PBS) was applied for2 h after permeabilization. Cultures were incubated with primary anti-body (1:100 anti-phospho-S6 (Ser235/236); Cell Signaling Technology,or 1:200 anti-MAP2 and 1:500 anti-SMI-312; Covance) overnight at 4°C,

washed, and incubated with appropriate secondary antibodies conju-gated with Alexa Fluor 488, 568, or 647 (Life Technologies) overnight.Cultures were then washed, counterstained with Nissl stain (only forp-S6), mounted, and imaged with a confocal scanning microscope (Leicaor Zeiss).

Western Blots and analysis. Cultures were homogenized, and proteinextracts were loaded into 12% Tris-Glycine polyacrylamide gels (Novexgels; Invitrogen) and separated by electrophoresis. Proteins were thentransferred to PVDF membranes (Millipore) and stained with a primaryantibody mixture (1:500 rabbit polyclonal anti-phospho-S6 (Ser 235/236) or anti-phospho-S6 (Ser 240/244), and 1:500 mouse monoclonalanti-S6, Cell Signaling Technology, in blocking buffer) followed by ap-propriate secondary antibodies from LI-COR. Membranes were imagedon a two-channel infrared scanner (Odyssey; LI-COR) and bands werequantified with LI-COR software.

Electrographic analysis. Ictal events (electrographic seizures) were de-fined as either paroxysmal events of much larger amplitude than back-ground multiple unit activity and lasting longer than 10 s, or shorterparoxysmal events that occurred with frequency of at least 2 Hz for atleast 10 s. Interictal events (events between seizures) were defined asparoxysmal events of �200 ms duration occurring at a frequency of �2Hz (Dyhrfjeld-Johnsen et al., 2010; Berdichevsky et al., 2012). In thosecases where ictal events were preceded or followed by interictal activity,start and end of an ictal event were defined as time points where parox-ysmal event frequency exceeded or decreased �2 Hz, respectively.

Axon sprouting quantification. We took phase contrast micrographs ofthe slice and surrounding area with sprouted axons at 20� magnifica-tion. Approximately 30 – 40 micrographs were needed to cover thesprouting area. We stitched the images together in Adobe PhotoshopCS2, using an automated batch stitching tool. We then traced the sliceoutline, and generated equidistant expanding contours from the sliceoutline at 200 pixel steps, using Fiji (extension of ImageJ software; Na-tional Institutes of Health, Bethesda, MD). Neurites crossing the con-tours were marked and counted using Cell Counter plug-in in Fiji. Weselected crossing counts at the contour at 900 �m distance from the edgeof the slice to quantify the number of sprouted axons, since sproutingdendrites are not long enough to extend so far from the neuronal soma(Taylor et al., 2005). Axon crossings were subdivided by the proximalhippocampal area: CA1 and subiculum, CA3, or dentate gyrus (DG).

Statistical analysis. We used Student’s t test for two-variable compari-sons, and one-way ANOVA with Holm–Sidak post hoc analysis for mul-tiple variable comparisons. For correlation analysis, we used PearsonProduct Moment to calculate correlation coefficients and p values.

ResultsLactate and LDH concentrations reflect ictal activity and ictalcell deathIn organotypic hippocampal cultures, interictal and ictal activi-ties follow a latent period of �7 d after dissection injury (Fig. 1A).We measured lactate and LDH concentrations in media fromorganotypic cultures that were chronically treated by 100 �Mphenytoin, 3 mM kynurenic acid (KYNA), or vehicle. Data from7, 10, 14, 17, 21, 24, 28, and 32 DIV were combined into fourtime points by arithmetic mean, DIV �{7,10}�, �{14,17}�,�{21,24}�, and �{28,32}� (Fig. 1 B,C). Lactate and LDH con-centrations in control (vehicle) cultures were significantly higherthan in cultures treated by either phenytoin or KYNA (n � 12cultures from three animals, each condition).

In epileptic control cultures, when ictal activity was present,lactate production was 1.88 � 0.38 mM/d (Fig. 1F; average � SD,n � 8 periods between medium changes, seven cultures), while inthe absence of ictal events, lactate was produced at 1.12 � 0.22mM/d (p � 0.001, t test; average � SD, n � 16 periods betweenmedium changes in seven slice cultures). Lactate production andduration of electrical activity were strongly correlated (Fig. 1D;n � 7 cultures, Pearson r � 0.622, p � 0.01), as were LDHconcentrations and PI counts (Fig. 1E; n � 6 cultures, data from

Berdichevsky et al. • PI3K-Akt-mTOR Signaling in Epilepsy J. Neurosci., May 22, 2013 • 33(21):9056 –9067 • 9057

Figure 1. Validation of lactate and LDH measurements as biomarkers for epileptic activity and cell death. A, Representative time course of epileptogenesis in an organotypic hippocampal culture.Colors correspond to the frequency of paroxysmal event occurrence in 10 s bins, with examples shown on the right. Deep blue, multiunit activity (top trace); light blue or yellow, interictal activity(middle trace); red, electrographic seizures (ictal events, bottom trace). B, C, time course of lactate (B) and LDH (C) concentration and cumulative production in epileptic controls (�) and culturestreated with phenytoin (‚), and KYNA (�); n � 12 each condition. Error bars indicate SEM. Statistical significance is shown for epileptic controls versus KYNA-treated cultures, with *p � 0.05,**p � 0.01, and ***p � 0.001. Statistically significant differences between epileptic controls and phenytoin-treated cultures were the same as control versus KYNA, except for lactate at DIV�{28,32}� and LDH at DIV �{21,24}� and �{28,32}�, where no statistically significant differences were found. D, Relationship of lactate concentration and activity, during time periods whenonly physiological or interictal activity was present (E) or when ictal activity was present (F). Lactate concentrations were plotted versus the duration of electrical activity, with each data pointcorresponding to the time between medium changes. Exponential fit line is shown. E, Relationship between LDH concentration in culture supernatant and the number of dead (PI-positive) cells. LDHconcentrations and numbers of PI-stained dead cells were measured in the same cultures, with each time point again corresponding to the time between medium changes (LDH accumulated in theculture medium since last medium change vs number of PI cells at the time of medium change). Linear fit line is shown. F, Lactate production (average � SD, ***p � 0.001) during time periods withand without ictal vents, from data shown in D. G, LDH release (average � SD, ***p � 0.001) from cultures during time periods with and without ictal events, from data in E.

9058 • J. Neurosci., May 22, 2013 • 33(21):9056 –9067 Berdichevsky et al. • PI3K-Akt-mTOR Signaling in Epilepsy

seven medium changes, r � 0.635, p � 0.001). LDH release dur-ing time periods when ictal events occurred was significantlyhigher than LDH release during periods with no ictal events (Fig.1G; p � 0.001, n � 6 cultures).

Manipulation of mTOR signaling: effects on ictal activity andictal cell deathWe tested the effects of inhibition of kinases (Table 1) on thedevelopment of epilepsy (Fig. 2A–F; n � 3 each condition). In-hibition of JNK had no significant effect, while inhibition of MEKhad no significant effect on lactate production before 20 DIV.However, inhibition of PI3K, Akt, mTOR, and dual inhibition ofPI3K-mTOR significantly decreased lactate production startingfrom 6 DIV (p � 0.05), and LDH release starting from 9 –10 DIV(p � 0.05). We repeated the experiments with LY 294002 andrapamycin (n � 3) and again found significant inhibition of lac-tate production (from 6 DIV, p � 0.05) and LDH release (from 10DIV, p � 0.05) relative to controls. We also tested the dose–response of API-2 and NVP-BEZ235 (n � 3 per condition), andfound dose-dependent inhibition of both lactate and LDH (Fig.3). We calculated lactate and LDH concentrations in drug-treated cultures as percentage of paired controls at the day of peaklactate concentration (maximum ictal activity; Fig. 2G). Signifi-cant differences were found between controls and rapamycin(n � 6, p � 0.001), 0.5 �M API-2 (n � 3, lactate p � 0.01, LDHp � 0.001), LY 294002 (n � 6, p � 0.001), and 40 nM NVP-BEZ235 (n � 3, p � 0.001).

We recorded electrographic activity in rapamycin, LY 294002,and control cultures between 10 and 15 DIV (Fig. 4A–E). Ictalactivity was present in �85% of treated and control cultures.However, there were significant 62 and 45% reductions in thenumber of ictal events per hour in rapamycin (n � 16) and LY294002-treated cultures (n � 14), respectively, versus controls(n � 13; p � 0.001). Ictal durations were unchanged (p � 0.21)between controls (36.7 � 3.3 s), rapamycin (32.9 � 3.8 s), and LY294002-treated cultures (28 � 2.4 s, average � SEM). The totaltime cultures spend in ictus per hour was significantly increasedin controls versus rapamycin and LY 294002-treated cultures(p � 0.001).

Pyramidal neuronal layers were well organized in control ep-ileptic cultures at 28 DIV (Fig. 4F); however, the number ofsurviving CA3c neurons at 28 DIV was �50% higher in culturestreated with either rapamycin or LY 294002 versus controls (Fig.4 G; p � 0.01, n � 3). The number of surviving CA1 neurons inrapamycin and LY 294002-treated cultures was �100% higherthan in controls (p � 0.01, n � 3). In a separate experiment, wefound similar numbers of surviving neurons in control cultures,but 70 –230% more surviving neurons in NVP-BEZ235-treatedcultures compared with controls in CA3c (p � 0.01, n � 3) and inCA3b (p � 0.05, n � 3).

Effect of rapamycin wash-in and wash-out onepileptic activityRapamycin treatment from 3 to 28 DIV had the same effect asdescribed above (n � 6 cultures from two additional experi-ments; Fig. 5A). Late application of rapamycin from 10 to 28 DIVresulted in significantly lower lactate concentrations comparedwith controls from 17 to 28 DIV, and in significantly lower LDHconcentrations from 17 to 24 DIV (n � 6; Fig. 5B). Early termi-nation of rapamycin treatment (rapamycin applied 3–17 DIV;Fig. 5C) did not increase the rate of subsequent lactate accumu-lation (1.19 � 0.314 mM/d on 17 DIV, 1.16 � 0.448 mM/d on 28DIV, average � SD, n � 6, p � 0.9) or LDH accumulation(0.29 � 0.2 a.u./d on 17 DIV, 0.29 � 0.16 a.u./d on 28 DIV,average � SD, n � 6, p � 0.97).

We did not see a significant change in the power of electro-graphic activity after 1 h of rapamycin wash-out followingchronic rapamycin treatment (14 —24 DIV) compared with elec-trographic activity in rapamycin (Fig. 5D,E; n � 3, p � 0.122,paired t test). We also did not see significant changes in timeseizing (duration of ictal activity per hour of recording) in cul-tures treated with rapamycin from 3 to 17 DIV and recorded on17 DIV versus cultures treated with rapamycin from 3 to 17 DIVand recorded after 11 d of wash-out (i.e., control conditions) on28 DIV (Fig. 5F; same experimental design as Fig. 5C; n � 18rapamycin-treated 17 DIV cultures and n � 19 28 DIV cultures inwhich rapamycin treatment was discontinued on 17 DIV (11 dwash), p � 0.447, Mann–Whitney rank sum test).

Chronic rapamycin treatment decreases axon sproutingWe quantified the number of sprouted axons at 8 DIV (Fig. 6).We confirmed that neurites crossing contours 900 �m away fromthe edge of organotypic culture were axons by staining for SMI-312 (axon marker; Ulfig et al., 1998; Chung et al., 2003) andMAP-2 (dendritic marker). Both of these markers stain only neu-ronal, and not glial, processes. Representative images are shownfor an area close to the edge of the slice (Fig. 6D), and an area 900�m away from the edge of the slice (Fig. 6E). Large numbers ofSMI-312-positive axonal processes, but only a few MAP-2-positive dendritic processes, leave stratum oriens and extend out-side of the slice (Fig. 6D), confirming that most of the sproutedprocesses outside of the slice are axons. Further away from theslice, all of the processes are positive for SMI-312, and none arepositive for MAP-2 (Fig. 6E), confirming that only sproutingaxons cross the 900 �m contour. Axon sprouting was signifi-cantly reduced in rapamycin-treated cultures compared withcontrols (Fig. 6F; average � SEM, n � 8 cultures, p � 0.05).Rapamycin treatment did not significantly reduce axon sprout-ing from CA1 and subiculum (p � 0.381), but did significantlyreduce sprouting from CA3 and DG areas (p � 0.01 in CA3, p �0.05 in DG; Fig. 6G). The lack of effect in area CA1 may reflect themuch higher neuronal density in this area in the rapamycin-treated cultures (Fig. 4F,G). Termination of rapamycin treat-ment for 11 d did not result in significant new axon sprouting(n � 8 cultures, comparison of rapamycin-treated cultures at 17DIV, and same cultures at 28 DIV after 11 d of rapamycin wash-out, p � 0.686, paired t test; Fig. 6H).

LY 294002 and rapamycin reduce phosphorylation of S6ribosomal proteinPhosphorylation of S6 (Ser 235/236) significantly increased onday 1 after dissection versus immediately after dissection (Fig. 7A;n � 4, p � 0.01). The ratio of pS6 to S6 then decreased by 7 DIV,and significantly increased again by 21 DIV (1.00 � 0.30 vs

Table 1. Inhibitor names, concentrations, and targets

Target name Target abbreviation Drug name Concentration

c-Jun N-terminal kinase JNK SP 600125 10 �MMAP kinase kinase MEK U0126 20 �MMammalian target of rapamycin mTOR Rapamycin 20 nMProtein kinase B Akt API-2 0.5 �MPhosphoinositide 3-kinase PI3K LY 294002 10 �MDual inhibitor PI3K-mTOR NVP-BEZ235 40 nM


1.88 � 0.43, 7 and 21 DIV, respectively, average � STD, n � 4,p � 0.05). Phosphorylation of S6 at Ser 235/236 was significantlyreduced in pyramidal cells in rapamycin and LY 294002-treatedcultures at 7 DIV compared with controls (Fig. 7 B; n � 3). Theratio of phospho-S6 (Ser 235/236) versus total S6 was reduced inrapamycin and LY 294002-treated cultures (94 and 81%, respec-tively, p � 0.001, n � 9, Fig. 7C). The ratio of phospho-S6 (Ser240/244) versus total S6 was also significantly reduced in rapa-mycin and LY 294002-treated cultures (99 and 91%, respectively,p � 0.001, n � 6; Fig. 7D).

DiscussionKey findingsWe found that lactate and LDH assays of culture media sam-pled at twice-weekly media changes provided useful estimatesof ictal activity and ictal cell death in organotypic hippocam-

pal cultures as they became epileptic in the first 10 DIV. PI3Kand mTOR inhibition reduced S6 phosphorylation, and PI3K,Akt, mTOR, and combined mTOR/PI3K inhibition reducedcell death, axon sprouting, and the severity of epilepsy. Wash-out experiments indicated that mTOR inhibition had no di-rect anticonvulsant effect, and that the anti-epileptogenic andneuroprotective effects of mTOR inhibition persisted for 11 dafter wash-out.

Lactate and LDH assaysIn the organotypic hippocampal culture model of pediatric post-traumatic epilepsy, interictal and ictal activities appear during thesecond week after initial injury (Dyhrfjeld-Johnsen et al., 2010).Ictal activity is accompanied by a significant increase in activity-dependent cell death, and both ictal activity and cell death areprevented by KYNA and phenytoin (Pozzo Miller et al., 1994;

Figure 2. Screen of inhibitors. A, SP 600125, JNK inhibitor. B, U0126, MEK inhibitor. C, Rapamycin, mTOR inhibitor. D, API-2, Akt inhibitor. E, LY 294002, PI3K inhibitor. F, NVP-BEZ235, dualPI3K/mTOR inhibitor. n � 3 per data point, for cultures with vehicle (F) and inhibitor treatment (E) lactate, and for cultures with vehicle (�) and inhibitor treatment (�) LDH. Cumulativeaverages of lactate and LDH production are plotted in each figure as solid gray line (vehicle-treated cultures) or dashed gray line (inhibitor-treated cultures). G, Summary of the data at the day of peaklactate production. N � 3 for both controls and treated slices for all groups except rapamycin (n � 6) and LY 294002 (n � 6). Error bars indicate SD.


Berdichevsky et al., 2012). If untreated, ictal activity persists forlonger than 4 weeks in vitro (Fig. 1A). The time course of epilep-togenesis in cultures from different animals varied slightly, sim-ilar to variances previously found in epileptogenesis in individualanimals subject to the same insult (Williams et al., 2009; Kadamet al., 2010). For this reason, all experiments were controlled bycomparing vehicle- and inhibitor-treated slices from the sameanimal.

We found increased lactate production due to ictal activity,and increased LDH release due to activity-dependent cell death.Ictal lactate accumulation occurs in humans (During et al., 1994;Castillo et al., 2001; Mercimek-Mahmutoglu et al., 2012) and inanimal models of epilepsy (Chapman et al., 1977; Fujikawa et al.,1988; Wasterlain et al., 2010). LDH release is a widely usedmarker of cell death (Decker and Lohmann-Matthes, 1988;Bonfoco et al., 1995). Measurements of lactate and LDH con-centrations in organotypic hippocampal cultures are fasterand less labor intensive than electrographic recordings anddead cell counts, enabling rapid analysis of chronic drug ef-fects including dose–response (Fig. 3). Lactate and LDH levelswere significantly reduced by phenytoin and KYNA acid,known inhibitors of ictal activity and cell death in organotypic

cultures (Berdichevsky et al., 2012). Lactate and LDH assayswere also congruent with electrographic and Nissl measure-ments of rapamycin and LY 294002 effects on epileptiformactivity and neuronal survival. These findings support the util-ity of lactate and LDH measurements as biomarkers of theprogression of epilepsy in this model.

Total lactate production by an organotypic hippocampal cul-ture depends on the number of surviving neurons in the culture.It has been previously shown that spontaneous seizures in orga-notypic cultures result in ongoing neural death (Pozzo Miller etal., 1994; Berdichevsky et al., 2012), leaving fewer neurons tomake lactate. This may result in decreased lactate productionsuch as that seen in vehicle-treated cultures in Figures 1B and2A–F between 3 and 4 weeks in vitro. To interpret whether thisdecrease corresponds to a decrease in epileptic activity or a de-crease in the number of neurons, LDH assays of cell death arenecessary. The time course of LDH production in vehicle-treatedepileptic cultures in Figures 1C and 2A–F shows significant in-creases in cell death after the end of latent period. Cell death dueto seizures is ongoing and cumulative (Fig. 1C), and by 28 DIVresults in a dramatic reduction in the number of surviving neu-rons (Fig. 4F,G; Berdichevsky et al., 2012). In these cultures,

Figure 3. Lactate and LDH concentrations show dose–response to varying concentrations of inhibitors. A, Dose–response to chronic API-2 treatment, n � 3 per point/condition. B, Dose–response to chronic NVP-BEZ235 treatment, n � 3 per point/condition. Concentrations are shown in the figure. Error bars indicate SD.


Figure 4. Electrical recordings and neuronal counts confirm that LY 294002 (LY) and rapamycin (RAPA) treatments reduce number of ictal events and promote neuronal survival. A, Typical ictalevent in an epileptic control culture. B, Representative electrical recordings from control, rapamycin, and LY 294002-treated cultures. Rapamycin and LY 294002 treatments reduce seizure frequency(C), but not seizure duration (D), with the overall effect of reducing the time spent in ictal state (E). **p � 0.01, ***p � 0.001. F, Representative images of Nissl-stained organotypic hippocampalcultures at 28 DIV. Scale bars, 50 �m. G, Neuronal counts revealed that more neurons survive in CA3c and CA1 of rapamycin and LY 294002-treated cultures compared with epileptic controls, n �3 per condition, **p � 0.01, ***p � 0.001.


lactate production is reduced due to lower number of neurons,but ictal activity has not ceased. In contrast, cultures treated withKYNA, phenytoin, or inhibitors of PI3K-Akt-mTOR pathwayhave reduced or no ictal activity, and correspondingly, no

seizure-dependent cell death (as shown by LDH assays). En-hanced neuron survival in these cultures (Figs. 4F,G) results inhigher rates of physiological lactate production, because there aremore neurons to make lactate. At later time points (fourth week

Figure 5. Effects of different time courses of rapamycin (RAPA) treatment on progression of epilepsy. A, Rapamycin is applied 3 to 28 DIV, 10 –28 DIV (B), 3–17 DIV (C). Control lactateconcentrations are plotted as F, while lactate concentrations in the rapamycin experimental group are plotted as E. LDH concentrations are plotted as � for control group, and � forrapamycin group. N � 6 per condition, per data point, *p � 0.05, **p � 0.01, ***p � 0.001. D, Representative electrical recording of a culture that was chronically treated withrapamycin, showing no change in activity upon rapamycin wash-out. E, Results of rapamycin washout in three cultures, E represents power of electrical activity in rapamycin, and Fpower of activity after wash-out. F, Comparison of seizure activity in rapamycin-treated cultures at 17 DIV (empty circle), and cultures in which rapamycin treatment was discontinuedstarting at 17 DIV for 11 d (filled circle).


in vitro), lactate production from a chronically epileptic slice withfewer surviving neurons may be the same as the rate of physio-logical lactate production in a treated slice. Thus, differences inlactate production diminish over time, and analysis of cell deathvia LDH assay is necessary for interpretation of lactate assay as amarker of ictal activity.

mTOR and epilepsyLactate and LDH levels were dramatically reduced once ictalevents began to occur (9 –17 DIV) in cultures treated with inhib-itors of PI3K, Akt, and mTOR. The reductions in lactate and LDHdue to rapamycin (mTOR), API-2 (Akt), and LY 294002 (PI3K)treatments were similar and confirmed by direct measurementsof electrographic activity and cell survival, which suggested thatPI3K-Akt signaling was responsible for most of mTOR activationas measured by physiological effects on epilepsy and cell death.The phosphorylation level of S6 ribosomal protein (a down-stream effector of mTOR pathway phosphorylated through S6K)is a widely used marker of mTOR activation (Guertin and Saba-tini, 2007; Buckmaster et al., 2009; Zeng et al., 2009). Phosphor-ylation of S6 has been reported to increase after epileptogenicbrain injury (Buckmaster et al., 2009; Zeng et al., 2009; Huang etal., 2010; Zhang and Wong, 2012) and hypoxic seizures (Talos etal., 2012). Interestingly, phosphorylation of S6 was biphasic inour experiments: a transient increase immediately after initialinjury (dissection) followed by a second increase during the pe-riod of ictal activity and ictal cell death. These data suggest thatinjury, whether caused by physical trauma or by ictal activity, canincrease mTOR activation. On the other hand, S6 remained mod-erately phosphorylated throughout the latent period, suggestingthat mTOR activation may play a role in circuit reorganizationbefore appearance of spontaneous seizures. Chronic LY 294002treatment reduced phosphorylation of S6 at 7 DIV (near the endof the pre-ictal period) as effectively as chronic rapamycin treat-ment, again suggesting that most of mTOR activation during thepre-ictal period occurs through PI3K. Finally, a dual PI3K-mTOR inhibitor NVP-BEZ235 (Maira et al., 2008) reduced lac-tate and LDH concentrations as much as PI3K or mTORinhibitors alone. The absence of additive or synergistic effectsconfirms that anti-epileptic effects of PI3K inhibition are due toinactivation of mTOR. Our findings demonstrate that mTORactivation in post-traumatic epileptogenesis is strongly PI3K de-pendent, suggesting that activation of PI3K by growth factorsmay be an important contributing factor to the development ofepilepsy after trauma. Considering the complexity and feedbackspresent in mTOR signaling (Guertin and Sabatini, 2007) andinvolvement of PI3K and Akt in mTOR activation, dual inhibi-tors such as NVP-BEZ235 or other drugs with similar mecha-nisms (Engelman, 2009; Borders et al., 2010) may provepreferable to rapamycin in treatment of epilepsy.

Anti-epileptogenic effects of mTOR inhibitionmTOR inhibition could be either anti-epileptogenic (preventingthe development of epilepsy), or merely anticonvulsant. Data

Figure 6. mTOR inhibition reduces axon sprouting. A, Left, Stitched micrographs show or-ganotypic hippocampal culture and surrounding area with sprouted axons. The inset showssprouting axons with quantification markers at different distances from the culture. B, C, Rep-resentative images of a control culture (B) and rapamycin-treated culture (C) are shown at 8DIV, with marks corresponding to axons crossing the contour 900 �m away from CA3 (blue), DG(green), or CA1 and subiculum (yellow) border. D, Representative SMI-312 (green) and MAP-2(red) staining of processes at the edge (white line) of an 8 DIV organotypic culture; right image

4

is merged SMI-312 and MAP-2 staining (stratum oriens, SO). E, Representative SMI-312 (green,left image) and MAP-2 (red, middle image), and merged SMI-312/MAP-2 (right image) stainingof processes 900 �m away from the edge of an 8 DIV organotypic culture. F, Comparison of thenumbers of axons crossing 900 �m contour in rapamycin or vehicle-treated cultures (n � 8,p � 0.05), G, Numbers of axons originating from different areas of the hippocampal culture,*p � 0.05. H, Comparison of the numbers of axons crossing 900 �m contour in rapamycin-treated cultures at 17 DIV (empty circles) versus axons in the same cultures 11 d after discon-tinuation of rapamycin treatment (filled circles).


from acute applications of rapamycin invitro do not suggest a direct anticonvul-sant effect (Daoud et al., 2007; Rüegg etal., 2007). Structural changes such as thereduction of mossy fiber sprouting afterchronic mTOR inhibition suggest an anti-epileptogenic effect (Buckmaster et al.,2009; Zeng et al., 2009; Huang et al.,2010). In our model, mTOR inhibitionwith rapamycin also led to reducedsprouting at 8 DIV. In the organotypichippocampal culture model of epilepto-genesis, 8 DIV is the latest time point atwhich this comparison can be made with-out confounding by the onset of seizuresand ictal cell death, which may produceadditional sprouting. Our results suggestthat inhibition of mTOR signaling afterinjury, but before onset of seizures, re-duces injury-induced axon sprouting.Suppression of mossy fiber sprouting re-quired ongoing treatment with rapamy-cin (Buckmaster et al., 2009), suggestingthat temporary mTOR inhibition doesnot have sustained anti-epileptogenic ef-fects. However, we found that after earlytermination of rapamycin treatment, lac-tate production, and LDH release re-mained suppressed for at least 11 d. It isimportant to note that at later time points,rapamycin-treated cultures produce lowlevels of lactate despite retaining manymore neurons than control (epileptic)cultures that have experienced significantictal cell death. Lactate production is sim-ilar in rapamycin-treated and vehicle-treated cultures at later time points(fourth week in vitro) due to significantneural death in vehicle-treated culturesfrom seizures, and enhanced neuronalsurvival in rapamycin-treated cultures, asdiscussed above. Low lactate productionfrom large numbers of neurons is consis-tent with strong attenuation of ictal activ-ity in rapamycin-treated cultures, evenafter rapamycin wash-out. Reductions inaxon sprouting and electrographic sei-zures persisted after rapamycin wash-outand confirmed results of lactate and LDHassays. These findings imply that in our

Figure 7. Phosphorylation of S6 ribosomal protein is reduced with chronic rapamycin or LY 294002 treatment. A, Changes in S6phosphorylation with time in vitro. Top: Representative Western blots of phospho-S6 (Ser 235/236) and total S6, from 0 to 7 DIV,

4

and from 7 to 27 DIV. Bottom, Ratios of phospho-S6 (Ser 235/236) to total S6 are plotted (n � 4, average �/� STD, *p �0.05, **p � 0.01). B, Immunohistochemistry micrographs ofcultures with different treatments, Nissl stain is shown in redand phosphor-S6 is in green. C, Representative Western blot ofpS6 (Ser235/236) and total S6 and comparison of pS6 (Ser235/236)/S6 ratio in cultures treated with vehicle, rapamycin,or LY 293002. D, Representative Western blot of pS6 (Ser 240/244) and total S6, and comparison of pS6 (Ser 240/244)/S6ratio in cultures treated with vehicle, rapamycin, or LY 294002(***p � 0.001).


model, transient inhibition of mTOR has sustained effects onepilepsy. Finally, it was reported that rapamycin reduces seizureactivity even when given after the development of spontaneousrecurrent seizures (Huang et al., 2010). Our findings were simi-lar: epileptiform activity and resulting cell death were both sig-nificantly reduced by late rapamycin treatment (starting on 10DIV) compared with epileptic controls (Fig. 5B), suggesting thatrapamycin was effective even in this relatively late stage of epilep-togenesis. While these findings may have clinical implications fortreatment of acquired epilepsy, they suggest that epileptogenesismay be a more dynamic process than previously imagined. Ictalcell death (Berdichevsky et al., 2012) may alter epileptic networksin a fashion that requires ongoing circuit plasticity to maintainepilepsy, and this ongoing epileptogenesis might be blocked bylate mTOR inhibition.

Utility and limitations of the organotypic slice preparationfor mTOR signalingHippocampal organotypic slice cultures provide an exceptionallyrapid model of epileptogenesis that retains key features of thisprocess (Dyhrfjeld-Johnsen et al., 2010; Berdichevsky et al., 2012;Sabolek et al., 2012). Because the slices are created at P7, this is amodel of epileptogenesis after pediatric brain injury, and some ofour results may be specific to the pediatric population. The slicecultures have microglia but few other elements of the immunesystem and no blood– brain barrier, so anti-inflammatory andvascular effects of mTOR inhibition can be largely excluded. Theaccessibility of the preparation lends itself to rapid assays of bothseizure activity and neuroprotection. The high throughput of thisapproach is particularly important for the exploration of the in-teraction of brain injury, epilepsy, and the effect of manipulationof complex signaling systems such as the mTOR pathway, inwhich not only the point of pathway intervention but also thetiming relative to injury are important variables.

ReferencesBerdichevsky Y, Dzhala V, Mail M, Staley KJ (2012) Interictal spikes, sei-

zures and ictal cell death are not necessary for post-traumatic epilepto-genesis in vitro. Neurobiol Dis 45:774 –785. CrossRef Medline

Bonfoco E, Krainc D, Ankarcrona M, Nicotera P, Lipton SA (1995) Apopto-sis and necrosis: two distinct events induced, respectively, by mild andintense insults with N-methyl-D-aspartate or nitric oxide/superoxide incortical cell cultures. Proc Natl Acad Sci U S A 92:7162–7166. CrossRefMedline

Borders EB, Bivona C, Medina PJ (2010) Mammalian target of rapamycin:biological function and target for novel anticancer agents. Am J HealthSyst Pharm 67:2095–2106. CrossRef Medline

Buckmaster PS, Ingram EA, Wen X (2009) Inhibition of the mammaliantarget of rapamycin signaling pathway suppresses dentate granule cellaxon sprouting in a rodent model of temporal lobe epilepsy. J Neurosci29:8259 – 8269. CrossRef Medline

Caron E, Ghosh S, Matsuoka Y, Ashton-Beaucage D, Therrien M, Lemieux S,Perreault C, Roux PP, Kitano H (2010) A comprehensive map of themTOR signaling network. Mol Syst Biol 6:453. Medline

Castillo M, Smith JK, Kwock L (2001) Proton MR spectroscopy in patientswith acute temporal lobe seizures. AJNR Am J Neuroradiol 22:152–157.Medline

Chapman AG, Meldrum BS, Siesjö BK (1977) Cerebral metabolic changesduring prolonged epileptic seizures in rats. J Neurochem 28:1025–1035.CrossRef Medline

Chiang GG, Abraham RT (2007) Targeting the mTOR signaling network incancer. Trends Mol Med 13:433– 442. CrossRef Medline

Chueh SC, Kahan BD (2005) Clinical application of sirolimus in renal trans-plantation: an update. Transpl Int 18:261–277. CrossRef Medline

Chung RS, Vickers JC, Chuah MI, West AK (2003) Metallothionein-IIApromotes initial neurite elongation and postinjury reactive neurite

growth and facilitates healing after focal cortical brain injury. J Neurosci23:3336 –3342. Medline

Crino PB (2011) mTOR: a pathogenic signaling pathway in developmentalbrain malformations. Trends Mol Med 17:734 –742. CrossRef Medline

Cronin J, Dudek FE (1988) Chronic seizures and collateral sprouting of dentatemossy fibers after kainic acid treatment in rats. Brain Res 474:181–184.CrossRef Medline

Daoud D, Scheld HH, Speckmann EJ, Gorji A (2007) Rapamycin: brainexcitability studied in vitro. Epilepsia 48:834 – 836. CrossRef Medline

Decker T, Lohmann-Matthes ML (1988) A quick and simple method for thequantitation of lactate dehydrogenase release in measurements of cellularcytotoxicity and tumor necrosis factor (TNF) activity. J Immunol Meth-ods 115:61– 69. CrossRef Medline

During MJ, Fried I, Leone P, Katz A, Spencer DD (1994) Direct measure-ment of extracellular lactate in the human hippocampus during sponta-neous seizures. J Neurochem 62:2356 –2361. Medline

Dyhrfjeld-Johnsen J, Berdichevsky Y, Swiercz W, Sabolek H, Staley KJ(2010) Interictal spikes precede ictal discharges in an organotypichippocampal slice culture model of epileptogenesis. J Clin Neuro-physiol 27:418 – 424. CrossRef Medline

Engelman JA (2009) Targeting PI3K signalling in cancer: opportunities,challenges and limitations. Nat Rev Cancer 9:550 –562. CrossRef Medline

Fujikawa DG, Vannucci RC, Dwyer BE, Wasterlain CG (1988) Generalizedseizures deplete brain energy reserves in normoxemic newborn monkeys.Brain Res 454:51–59. CrossRef Medline

Guertin DA, Sabatini DM (2007) Defining the role of mTOR in cancer.Cancer Cell 12:9 –22. CrossRef Medline

Huang X, Zhang H, Yang J, Wu J, McMahon J, Lin Y, Cao Z, Gruenthal M, HuangY (2010) Pharmacological inhibition of the mammalian target of rapamy-cin pathway suppresses acquired epilepsy. Neurobiol Dis 40:193–199.CrossRef Medline

Kadam SD, White AM, Staley KJ, Dudek FE (2010) Continuous electroen-cephalographic monitoring with radio-telemetry in a rat model of peri-natal hypoxia-ischemia reveals progressive post-stroke epilepsy.J Neurosci 30:404 – 415. CrossRef Medline

Lew FH, Buckmaster PS (2011) Is there a critical period for mossy fibersprouting in a mouse model of temporal lobe epilepsy? Epilepsia 52:2326 –2332. CrossRef Medline

Maira SM, Stauffer F, Brueggen J, Furet P, Schnell C, Fritsch C, Brachmann S,Chène P, De Pover A, Schoemaker K, Fabbro D, Gabriel D, Simonen M,Murphy L, Finan P, Sellers W, García-Echeverría C (2008) Identifica-tion and characterization of NVP-BEZ235, a new orally available dualphosphatidylinositol 3-kinase/mammalian target of rapamycin inhibitorwith potent in vivo antitumor activity. Mol Cancer Ther 7:1851–1863.CrossRef Medline

McDaniel SS, Wong M (2011) Therapeutic role of mammalian target ofrapamycin (mTOR) inhibition in preventing epileptogenesis. NeurosciLett 497:231–239. CrossRef Medline

Mercimek-Mahmutoglu S, Horvath GA, Coulter-Mackie M, Nelson T, Wa-ters PJ, Sargent M, Struys E, Jakobs C, Stockler-Ipsiroglu S, Connolly MB(2012) Profound neonatal hypoglycemia and lactic acidosis caused bypyridoxine-dependent epilepsy. Pediatrics 129:e1368 –1372. CrossRefMedline

Okazaki MM, Evenson DA, Nadler JV (1995) Hippocampal mossy fibersprouting and synapse formation after status epilepticus in rats: visualiza-tion after retrograde transport of biocytin. J Comp Neurol 352:515–534.CrossRef Medline

Pozzo Miller LD, Mahanty NK, Connor JA, Landis DM (1994) Spontaneouspyramidal cell death in organotypic slice cultures from rat hippocampus isprevented by glutamate receptor antagonists. Neuroscience 63:471–487.CrossRef Medline

Raymont V, Salazar AM, Lipsky R, Goldman D, Tasick G, Grafman J (2010)Correlates of posttraumatic epilepsy 35 years following combat brain in-jury. Neurology 75:224 –229. CrossRef Medline

Rüegg S, Baybis M, Juul H, Dichter M, Crino PB (2007) Effects of rapamycinon gene expression, morphology, and electrophysiological properties ofrat hippocampal neurons. Epilepsy Res 77:85–92. CrossRef Medline

Sabolek HR, Swiercz WB, Lillis KP, Cash SS, Huberfeld G, Zhao G, Ste MarieL, Clemenceau S, Barsh G, Miles R, Staley KJ (2012) A candidate mech-anism underlying the variance of interictal spike propagation. J Neurosci32:3009 –3021. CrossRef Medline

Salazar AM, Jabbari B, Vance SC, Grafman J, Amin D, Dillon JD (1985)


http://dx.doi.org/10.1016/j.nbd.2011.11.001http://www.ncbi.nlm.nih.gov/pubmed/22115940http://dx.doi.org/10.1073/pnas.92.16.7162http://www.ncbi.nlm.nih.gov/pubmed/7638161http://dx.doi.org/10.2146/ajhp100020http://www.ncbi.nlm.nih.gov/pubmed/21116000http://dx.doi.org/10.1523/JNEUROSCI.4179-08.2009http://www.ncbi.nlm.nih.gov/pubmed/19553465http://www.ncbi.nlm.nih.gov/pubmed/21179025http://www.ncbi.nlm.nih.gov/pubmed/11158901http://dx.doi.org/10.1111/j.1471-4159.1977.tb10665.xhttp://www.ncbi.nlm.nih.gov/pubmed/16986http://dx.doi.org/10.1016/j.molmed.2007.08.001http://www.ncbi.nlm.nih.gov/pubmed/17905659http://dx.doi.org/10.1111/j.1432-2277.2004.00039.xhttp://www.ncbi.nlm.nih.gov/pubmed/15730485http://www.ncbi.nlm.nih.gov/pubmed/12716941http://dx.doi.org/10.1016/j.molmed.2011.07.008http://www.ncbi.nlm.nih.gov/pubmed/21890410http://dx.doi.org/10.1016/0006-8993(88)90681-6http://www.ncbi.nlm.nih.gov/pubmed/3214710http://dx.doi.org/10.1111/j.1528-1167.2006.00976.xhttp://www.ncbi.nlm.nih.gov/pubmed/17284292http://dx.doi.org/10.1016/0022-1759(88)90310-9http://www.ncbi.nlm.nih.gov/pubmed/3192948http://www.ncbi.nlm.nih.gov/pubmed/8189240http://dx.doi.org/10.1097/WNP.0b013e3181fe0709http://www.ncbi.nlm.nih.gov/pubmed/21076333http://dx.doi.org/10.1038/nrc2664http://www.ncbi.nlm.nih.gov/pubmed/19629070http://dx.doi.org/10.1016/0006-8993(88)90802-5http://www.ncbi.nlm.nih.gov/pubmed/3136858http://dx.doi.org/10.1016/j.ccr.2007.05.008http://www.ncbi.nlm.nih.gov/pubmed/17613433http://dx.doi.org/10.1016/j.nbd.2010.05.024http://www.ncbi.nlm.nih.gov/pubmed/20566381http://dx.doi.org/10.1523/JNEUROSCI.4093-09.2010http://www.ncbi.nlm.nih.gov/pubmed/20053921http://dx.doi.org/10.1111/j.1528-1167.2011.03315.xhttp://www.ncbi.nlm.nih.gov/pubmed/22092282http://dx.doi.org/10.1158/1535-7163.MCT-08-0017http://www.ncbi.nlm.nih.gov/pubmed/18606717http://dx.doi.org/10.1016/j.neulet.2011.02.037http://www.ncbi.nlm.nih.gov/pubmed/21354266http://dx.doi.org/10.1542/peds.2011-0123http://www.ncbi.nlm.nih.gov/pubmed/22529283http://dx.doi.org/10.1002/cne.903520404http://www.ncbi.nlm.nih.gov/pubmed/7721998http://dx.doi.org/10.1016/0306-4522(94)90544-4http://www.ncbi.nlm.nih.gov/pubmed/7891859http://dx.doi.org/10.1212/WNL.0b013e3181e8e6d0http://www.ncbi.nlm.nih.gov/pubmed/20644150http://dx.doi.org/10.1016/j.eplepsyres.2007.09.009http://www.ncbi.nlm.nih.gov/pubmed/17983731http://dx.doi.org/10.1523/JNEUROSCI.5853-11.2012http://www.ncbi.nlm.nih.gov/pubmed/22378874

Epilepsy after penetrating head injury. I. Clinical correlates: a report of theVietnam Head Injury Study. Neurology 35:1406 –1414. CrossRef Medline

Sarbassov DD, Ali SM, Sabatini DM (2005) Growing roles for the mTORpathway. Curr Opin Cell Biol 17:596 – 603. CrossRef Medline

Sliwa A, Plucinska G, Bednarczyk J, Lukasiuk K (2012) Post-treatment withrapamycin does not prevent epileptogenesis in the amygdala stimulationmodel of temporal lobe epilepsy. Neurosci Lett 509:105–109. CrossRefMedline

Staley BA, Vail EA, Thiele EA (2011) Tuberous sclerosis complex: diagnosticchallenges, presenting symptoms, and commonly missed signs. Pediatrics127:e117–125. CrossRef Medline

Sutula T, Cascino G, Cavazos J, Parada I, Ramirez L (1989) Mossy fibersynaptic reorganization in the epileptic human temporal lobe. Ann Neu-rol 26:321–330. CrossRef Medline

Talos DM, Sun H, Zhou X, Fitzgerald EC, Jackson MC, Klein PM, Lan VJ,Joseph A, Jensen FE (2012) The interaction between early life epilepsyand autistic-like behavioral consequences: a role for the mammalian tar-get of rapamycin (mTOR) pathway. PLoS One 7:e35885. CrossRefMedline

Tang H, Long H, Zeng C, Li Y, Bi F, Wang J, Qian H, Xiao B (2012) Rapamycinsuppresses the recurrent excitatory circuits of dentate gyrus in a mouse model oftemporal lobe epilepsy. Biochem Biophys Res Commun 420:199–204. CrossRefMedline

Taylor AM, Blurton-Jones M, Rhee SW, Cribbs DH, Cotman CW, Jeon NL(2005) A microfluidic culture platform for CNS axonal injury, regener-ation and transport. Nat Methods 2:599 – 605. CrossRef Medline

Ulfig N, Nickel J, Bohl J (1998) Monoclonal antibodies SMI 311 and SMI312 as tools to investigate the maturation of nerve cells and axonal pat-terns in human fetal brain. Cell Tissue Res 291:433– 443. CrossRefMedline

van Vliet EA, Forte G, Holtman L, den Burger JC, Sinjewel A, de Vries HE,Aronica E, Gorter JA (2012) Inhibition of mammalian target of rapamy-cin reduces epileptogenesis and blood-brain barrier leakage but not mi-croglia activation. Epilepsia 53:1254 –1263. CrossRef Medline

Wasterlain CG, Thompson KW, Suchomelova L, Niquet J (2010) Brain en-ergy metabolism during experimental neonatal seizures. Neurochem Res35:2193–2198. CrossRef Medline

Williams PA, White AM, Clark S, Ferraro DJ, Swiercz W, Staley KJ, Dudek FE(2009) Development of spontaneous recurrent seizures after kainate-induced status epilepticus. J Neurosci 29:2103–2112. CrossRef Medline

Zeng LH, Rensing NR, Wong M (2009) The mammalian target of rapamy-cin signaling pathway mediates epileptogenesis in a model of temporallobe epilepsy. J Neurosci 29:6964 – 6972. CrossRef Medline

Zhang B, Wong M (2012) Pentylenetetrazole-induced seizures causeacute, but not chronic, mTOR pathway activation in rat. Epilepsia53:506 –511. CrossRef Medline


http://dx.doi.org/10.1212/WNL.35.10.1406http://www.ncbi.nlm.nih.gov/pubmed/3929158http://dx.doi.org/10.1016/j.ceb.2005.09.009http://www.ncbi.nlm.nih.gov/pubmed/16226444http://dx.doi.org/10.1016/j.neulet.2011.12.051http://www.ncbi.nlm.nih.gov/pubmed/22227620http://dx.doi.org/10.1542/peds.2010-0192http://www.ncbi.nlm.nih.gov/pubmed/21173003http://dx.doi.org/10.1002/ana.410260303http://www.ncbi.nlm.nih.gov/pubmed/2508534http://dx.doi.org/10.1371/journal.pone.0035885http://www.ncbi.nlm.nih.gov/pubmed/22567115http://dx.doi.org/10.1016/j.bbrc.2012.02.143http://www.ncbi.nlm.nih.gov/pubmed/22414694http://dx.doi.org/10.1038/nmeth777http://www.ncbi.nlm.nih.gov/pubmed/16094385http://dx.doi.org/10.1007/s004410051013http://www.ncbi.nlm.nih.gov/pubmed/9477300http://dx.doi.org/10.1111/j.1528-1167.2012.03513.xhttp://www.ncbi.nlm.nih.gov/pubmed/22612226http://dx.doi.org/10.1007/s11064-010-0339-4http://www.ncbi.nlm.nih.gov/pubmed/21136154http://dx.doi.org/10.1523/JNEUROSCI.0980-08.2009http://www.ncbi.nlm.nih.gov/pubmed/19228963http://dx.doi.org/10.1523/JNEUROSCI.0066-09.2009http://www.ncbi.nlm.nih.gov/pubmed/19474323http://dx.doi.org/10.1111/j.1528-1167.2011.03384.xhttp://www.ncbi.nlm.nih.gov/pubmed/22242835

PI3K-Akt Signaling Activates mTOR-Mediated Epileptogenesis in Organotypic Hippocampal Culture Model of Post-Traumatic EpilepsyIntroductionMaterials and MethodsResultsEffect of rapamycin wash-in and wash-out on epileptic activityChronic rapamycin treatment decreases axon sproutingDiscussionKey findingsLactate and LDH assays

mTOR and epilepsyAnti-epileptogenic effects of mTOR inhibitionReferences

Documents

NeurobiologyofDisease PI3K-AktSignalingActivatesmTOR ...traumatic epileptogenesis remain largely unknown (McDaniel and Wong, 2011), and effects of mTOR inhibitors on epilepsy have